Ethanol cooling a ghetto loop.

[GruntBlender]

Having just upgraded to an R9 290, my PC sounds like a jet, to the point where people behind walls are complaining. This won't do, so I'm going to to liquid cool the beast to reduce noise (and hopefully temperature.)
Now, I'm an idiot, engineer, genius, scientist, with a controversial dash of arrogance, so I decided to ignore common practice. Being poor, the loop will have the cheapest parts eBay has to offer (modified where necessary). Being mad, and this being my first liquid cooled setup, I've decided to go with ethanol instead of water. This means that in case of a leak, I don't have to worry about shorting, and if there's a very slow leak it'll evaporate with no problems.
I'm aware of the lower boiling point of ethanol, but if the temp gets to that anywhere in the loop, it'll mean bigger problems than just that. I'm also aware of the lower (about half) heat capacity of ethanol, and that's where I might run into problems. At worst, I'd be looking at double the DeltaT compared to water, but heat capacity isn't the only determinant of DeltaT.
I'll be using 2 240L/H pumps running in parallel to ensure redundancy in case of cheap components failing on me. 240L/H =~ 67ml/s, at least according to what's on the box. I'll be happy with 30mL from the parallel setup. Now, I'm planning on dumping around 300W through this system, so I'll have around 10J at most being transferred by each ml. If that were the only factor, that would give me around 3 degrees DeltaT with water and around 5 with ethanol. Not bad either way.
Ethanol also has about a quarter the thermal conductivity of water, but I'm hoping forced convection plays a much larger role in this sort of system than conduction does. With any luck I'll only see a few degrees of difference between what this system does and what it could do with water. Time will tell (eBay orders can take a month to arrive here).

2x 240L/H, 3m max head, water pumps running parallel for redundancy.
2x 40x40mm, $5 aluminium blocks for CPU & GPU, probably have to sand the bottom to get good contact.
200ml acrylic reservoir (easy to change if ethanol misbehaves with it chemically)
Silicone tubing, PVC fittings, other assorted junk.
120mm radiator (I know, I know. If it tests to be insufficient I have an old ATF cooler radiator lying about as backup).

Ethanol pros:
- No corrosion
- Non-conductive
- Quick evaporation
Cons:
- Low heat capacity
- Low conductivity
- Potential chemical interaction with acrylic components

I guess I'm posting this so you all can laugh at my naivete, and in hopes that someone might have experience with this sort of heresy and could give me a bit of advice or point out what I've missed.

 

[sizzling]

Please keep us updated, I'd love to know how you get on.

 

[Dunlop0078]

Lol sounds interesting, I would also like to know how this turns out. I have heard bad things about cheap ebay compression fittings, but if your not worried about leaking or shorting go for it.

 

[Paperdoc]

I have two serious reservations about this. Certainly one is the heat capacity issue. Using ethanol in your loop will reduce your heat removal rate substantially.

But my big concern is FIRE!! Ethanol is a a highly flammable liquid. Any spill or any leak will have liquid ethanol flowing in unpredictable places around a system with lots of hot spots. If it ignites, the fire will burn through your tubing, releasing ALL of that ethanol to feed the flames. Your entire system will burn up, and you MIGHT start a fie large enough to proceed to the entire room and house. I REALLY think this is NOT worth the risk!

 

[GruntBlender]

I have two serious reservations about this. Certainly one is the heat capacity issue. Using ethanol in your loop will reduce your heat removal rate substantially.

But my big concern is FIRE!! Ethanol is a a highly flammable liquid. Any spill or any leak will have liquid ethanol flowing in unpredictable places around a system with lots of hot spots. If it ignites, the fire will burn through your tubing, releasing ALL of that ethanol to feed the flames. Your entire system will burn up, and you MIGHT start a fie large enough to proceed to the entire room and house. I REALLY think this is NOT worth the risk!

Thank you for your reply.
With capacity, I think I'll have enough flow rate for it to not be a problem. It certainly will be a bit less effective, but I don't think capacity is the bottleneck.
Fire is a bit more worrisome. Ethanol does have a very low flash point, around room temperature. But the autoignition point is above 300C, I doubt any part of my PC will get that hot. I suppose sparks would be a source of ignition if there's large enough vapor pressure of ethanol, but that's more of an explosion risk than fire... Hmmm....
Ok, you have a good point, but I'll soldier on with my madness. I'll be banking on decent ventilation to remove fumes from leaks, and lack of sparks to reduce the chance of explosion. Anyway, I'm no exactly expecting enough leakage to be a fire hazard.

 

[GruntBlender]

Quick update, most of the parts arrived and they're better quality than I expected. The radiator impedes both coolant and air flow more than I expected.
Coolant shouldn't be a problem, I'll be running the two pumps in series. On paper they have 3m head, I crudely tested to over 1.5 each.
The bottleneck will probably be with airflow through the radiator. Tested with the fan I had and it's terrible. I ordered a couple more fans, I'll run them in push-pull config. Hopefully that gets me at least 40 or so cfm.

 

[synphul]

Ethanol sounds like a bit of a scary proposition due to flammability. Parts on a pc may not be 300c but there is a chance of spark. Not to mention the gpu backplate can often get pretty hot and it's right below the cpu waterblock. I can appreciate the desire of avoiding short risks with water though distilled water is pretty inert in terms of shorting. Still not a good idea to have leaks regardless.

There may be a corrosion risk of the tubing or pump internals running ethanol if they're not designed to handle it, eating away at the tubing (which would promote leaking) or damaging pump seals and causing premature pump failure.

Even if the pc itself may not be a risk, if anything else were to somehow cause a flame or catch fire and move to the pc, igniting the ethanol could be a bad day. Fingers crossed everything works out well, I just have safety concerns. Even without catching fire, ethanol boils at less than 80c which can be a somewhat common component temp. If it's a sealed system, what's going to happen when the ethanol begins to boil and pressure builds in the system? Even if it doesn't fail with a 'bang' the increased pressure is going to strain the fittings, pump, rad etc.

Just further things to consider.

 

[The Paladin]

This was his last post after testing, CNN reported 1 man dies last night in a massive apartment explosion.. neighbourghs called him nice, but eccentric.

 

[GruntBlender]

I'm going to stress test it with the reservoir open. If anything gets close to boiling, I'll switch to water. I'm not worried about ethanol eroding anything other than the reservoir. If that happens but the temps are ok, I'll make a glass one.
I don't have a back plate on my gfx card, not sure if I need one. Any advice?
The radiator is 17 fpi, BTW, so I'll probably have to wait for more fans to arrive.
Thank you for your concern, everyone.

Edit:
I made a tool to help figure out liquid cooling loop:
Github page
Download
It's a python script, takes pump and fan specs and gives you approximate deltas. It's nowhere near a definitive simulation, but a decent ballpark, I think. It's using laws of thermodynamics and some real world data I found regarding radiator and fan performance.

With my planned setup it gives me a difference of 2.5C in fluid temperature, indicating the pump and coolant are sufficient for my application. For the radiator though I get around 20 degree air heating to dump my 300W into the atmosphere, so the minimum coolant temp will be at least 20 degrees above ambient. I know it's not great, but for my purpose that's fantastic. I'll be happy with anything lower than 40 degree delta T on my GPU.
BTW, the Soviet Maxim variant was ethanol cooled. If it's good enough for the Red Army, it's good enough for me.

Current cooling performance:
Ambient = 22C
at 65% GPU fan speed (3670RPM): CPU 61C (dT 39), GPU 88C (dT 66)
at 100% GPU fan speed (5400RPM): CPU 58C (dT 36), GPU 72C (dT 50)
Even at 65%, the GPU fan is far too obnoxiously loud. I'm still waiting on some parts, this is going to be great.

 

[petriedish81]

I think your biggest problem is going to be evaporation. Ethanol seems like a dumb idea when we know water works great and there are so many caveats and dangers added by using ethanol. There are reasons for water being the coolant of choice. If you keep an eye on your loops you will be fine with water. A leak usually is apparent from the beginning.

Have you thought of using an all-in-one cooler or aftermarket air cooler? Both should be cheaper and much quieter. The NZXT G10 bracket paired with a cheap AIO cooler would be great for you. I have a custom cooling loop. It is a lot of money and effort when not really necessary. If I were to do it over again I would have gone AIO.

 

[GruntBlender]

I think your biggest problem is going to be evaporation. Ethanol seems like a dumb idea when we know water works great and there are so many caveats and dangers added by using ethanol. There are reasons for water being the coolant of choice. If you keep an eye on your loops you will be fine with water. A leak usually is apparent from the beginning.

Have you thought of using an all-in-one cooler or aftermarket air cooler? Both should be cheaper and much quieter. The NZXT G10 bracket paired with a cheap AIO cooler would be great for you. I have a custom cooling loop. It is a lot of money and effort when not really necessary. If I were to do it over again I would have gone AIO.

Yeah, I can't really splurge on an all-in-one or aftermarket air kit for the 290. The ones I saw have no cooling on the vram or vrm anyway. There shouldn't be evaporation since it's a closed loop. As long as I don't get the coolant above about 70C there shouldn't be a problem.
Water works great, but there are issues with corrosion and bacteria/algae growth, aren't there?

 

[The Paladin]

we know water works well, but we know " automobile coolant" works better, it is possible ethanol works great, but the flammable state of such a liquid just tells my tingly senses , this experiment can only end in a disaster... (ie a leak forms and your ethanol evaporate on hardware.. shivers thinking of the consequences... )

you are a brave man....I would not do it.

have you completed the rig yet ? I would love to see pictures.

 

[Venom Six Actual]

*Furiously looks though logs of customer data for OP's username*

 

[GruntBlender]

we know water works well, but we know " automobile coolant" works better, it is possible ethanol works great, but the flammable state of such a liquid just tells my tingly senses , this experiment can only end in a disaster... (ie a leak forms and your ethanol evaporate on hardware.. shivers thinking of the consequences... )

you are a brave man....I would not do it.

have you completed the rig yet ? I would love to see pictures.

Aye, a brave fool I am. I'm still waiting for a couple fittings, I'll definitely post the build and the final result, along with the results of a burn in run.

 

[GruntBlender]

An album of the parts I have. Not pictured is a bunch of 8mm hose. Still missing some 8mm barb fittings and half the VRAM heatsinks I want. http://imgur.com/a/wCX60

 

[the nerd 389]

Just so you're aware, ethanol is a ridiculously powerful solvent, especially with regards to certain plastics. It will eat most flexible hoses. It's not as bad as gasoline, though.

 

[GruntBlender]

Just so you're aware, ethanol is a ridiculously powerful solvent, especially with regards to certain plastics. It will eat most flexible hoses. It's not as bad as gasoline, though.

I've worked with it for a while, even gasoline is not as bad as people think. I'll test a bit of hose for peace of mind, but I'm sure it's fine. These vinyl hoses stand up to DOT-4 for a good while, and I've used them as low pressure fuel lines with no problems.

 

[the nerd 389]

In my experience, using the wrong plastic for a fuel line (lawnmower), it can last one or two weeks.

I.E. long enough to think everything is okay.

There's a reason they don't put beer in plastic, even the bottom shelf stuff.

 

[GruntBlender]

In my experience, using the wrong plastic for a fuel line (lawnmower), it can last one or two weeks.

I.E. long enough to think everything is okay.

There's a reason they don't put beer in plastic, even the bottom shelf stuff.

Not beer, no. But box wine, cheap cider, even methylated spirits are kept in plastic containers. PET is a hell of a plastic. Like I said, vinyl tubing can stand up to gasoline for years, ethanol is nowhere near as bad. Before it starts going bad, it hardens, so checking the tube every few months is just fine. Besides, there's practically no mechanical stresses in the system.

TL;DR: She'll be right, mate.

 

[GruntBlender]

I've been redoing some calculations with a python script, having looked up waterblock performance, and i'm getting some slightly less than fun figures for full power performance:

T1: Approx ?T from GPU to block: 46.75°C
T2: Approx ?T from CPU to block: 4.25°C
Ta: Min ?T from components to coolant: 2.71°C
Tb: Min ?T for coolant temperature: 16.33°C
Flow: 57.4 ml/s. 86% of max.
Pressure at pump: 0.278 m H²O

Approx GPU Temperature: 85.79°C
Approx CPU Temperature: 43.29°C

All that at 20C ambient. The largest issue is the GPU waterblock performance, apparently. I looked up real world data and modelled it in my script. I assumed the 275W TDP for the GPU, though I know it's going to be a bit lower.

All in all, this is shaping up to be an amazing learning experience for me.

Note: That GPU temp is above the boiling point of Ethanol. If any part of the inside of the water block gets that hot (fairly likely) I'm going to have either a problem or a solution. Either:
1. The boiling Ethanol will increase the pressure inside the loop to unworkable levels and I'll have to change to water, or
2. The boiling Ethanol will act like an evaporative cooler and quickly condense back, causing something like cavitation, and increasing the performance of the water block, giving me a few extra degrees of cooling. If so, I might design a custom block for the GPU to take advantage of that effect.

P.S., the data I used didn't specifically state the power transferred from the CPU to the water block, but indicated an i7-920 @4Ghz @1.40V which I guesstimated with the help of http://outervision.com/power-supply-calculator to be 247W. Is that in the ballpark?

 

[the nerd 389]

If you end up with two phase flow, you should include the heat of vaporization in your calculations. You could also increase your pump speed to minimize the amount of flashing that takes place.

It may also be worth considering designing that part of the loop as a vapor-compression chiller. You'd need to use a piston style pump, but the loop would be more efficient.

Lastly, you should have a PSV or blowoff valve if you're going to run the loop nearly that warm. If you don't, it could explode.

Regarding the CPU power, for reference, I'm running an i7-930 at 1.224V and 3.8 GHz. It produces less than it's rated TDP under load. I'm seeing about 112W peak in a stress test. Those CPUs are from the golden age of overclocking.

 

[GruntBlender]

If you end up with two phase flow, you should include the heat of vaporization in your calculations. You could also increase your pump speed to minimize the amount of flashing that takes place.

It may also be worth considering designing that part of the loop as a vapor-compression chiller. You'd need to use a piston style pump, but the loop would be more efficient.

Lastly, you should have a PSV or blowoff valve if you're going to run the loop nearly that warm. If you don't, it could explode.

Regarding the CPU power, for reference, I'm running an i7-930 at 1.224V and 3.8 GHz. It produces less than it's rated TDP under load. I'm seeing about 112W peak in a stress test. Those CPUs are from the golden age of overclocking.

Cheers.
I'm still expecting the coolant to average around 40C in the loop, so hopefullt I wouldn't need that much extra gear even if it is boiling at certain points. If the full 275W of the GPU would go into vaporising the coolant (A bloody odd phrase in a watercooling discussion) I'd be expecting around 200ml of gas per second at around atmospheric pressure. I might need to include an expansion chamber. I'll run it open and see if I get any problems during burn-in.
I was planning to go pump-CPU-GPU-radiator, but if there's gas coming out of the GPU I'd have to either route that straight into the radiator or include a condenser between the GPU and CPU.
Now there's an idea, buy a better block for the GPU and modify the current one with an expansion chamber to use as a condenser... IF there are problems with the setup as planned.

 

[the nerd 389]

You should consider running the setup under a little bit of pressure. If memory serves, you can bump the boiling point of ethanol up by quite a bit with just a few PSI. The radiators are rated up to 80 PSI, so you have a significant safety factor to work with. Just make sure you get the fittings in well enough.

 

[GruntBlender]

You should consider running the setup under a little bit of pressure. If memory serves, you can bump the boiling point of ethanol up by quite a bit with just a few PSI. The radiators are rated up to 80 PSI, so you have a significant safety factor to work with. Just make sure you get the fittings in well enough.

Boiling point goes up by about 2-2.5K per meter of water head. If there's boiling, it should reach that pressure quite quickly without artificial pressurisation. Thanks for pointing that out. Still, I'm not sure how much faith I have in the cheap components I'm using. The waterblocks are about 1.5mm aluminium. Pumps are even worse. Oh well, if it explodes, I'll be a lesson to others at least.

 

[the nerd 389]

Just remember what PSI means. It's 1 lbf per square inch. If the waterblock is 4 square inches, and you run it at 3 PSI, it has to support 12 lbf. That's not asking much from 1.5mm aluminum.